Brain tumor MRI image classification with feature selection and extraction using linear discriminant analysis

Rathi, V. P. Gladis Pushpa; Palani, S.

doi:10.48550/arxiv.1208.2128

Cited by 7 publications

(6 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After selecting the most relevant features, the following three classifiers were analyzed: LDA [77], Random Forest [64], and Logistic Regression [67]. An LDA classifier maximizes the ratio between-class variance to the within-class variance creating a separable decision boundary between the given classes [78]. Class conditional densities are then fit to the data using Naïve Bayes.…”

Section: Glomerular Classificationmentioning

confidence: 99%

Computational Classification Methods in Renal Pathology

Basso

2024

Preprint

View full text Add to dashboard Cite

When assessing kidney biopsies, pathologists use Light microscopy (LM), immunofluorescence, and electron microscopy to describe and diagnose glomerular lesions and diseases. These are laborious, costly, fraught with inter-observer variability, and can have delays in turn-around time. Thus, computational approaches can be designed as screening and/or diagnostic tools, potentially relieving pathologist time, healthcare resources, while also having the ability to identify novel biomarkers, including subvisual features. This thesis compares a novel Biomarker feature extraction (BFE) model and test several pre- trained deep learning models to diagnose three glomerular diseases using Periodic acid-Schiff (PAS) stained images. The BFE model extracted a panel of 233 explainable biomarkers related to colour, morphology, and microstructural texture that describe underlying pathology. These biomarkers were subsequently narrowed to ten morphological and microstructural texture fea- tures for classification. The proposed pre-trained deep learning models used data augmentation and Gradient-weighted Class Activation Mapping (Grad-CAM) for better performance and inter- pretability. All classification models were then used to classify Minimal change disease (MCD), Membranous nephropathy (MN), and Thin-basement membrane nephropathy (TBMN) diseases on a glomerular and patient-level basis. The BFE model resulted in a glomerular validation accuracy of 67.6% and testing accuracy of 76.8%. All deep learning approaches had higher validation accuracies (most for VGG16 at 78.5%) but lower testing accuracies. The highest testing accuracy at the glomerular level was VGG16 at 71.9%, while that at the patient-level was InceptionV3 at 73.3%. Patient-level classification results showed higher accuracy for the BFE model (86.76%) compared to all deep learning methods. The BFE model was superior to three established deep learning models. The small dataset was likely contributory to the lower performance of the deep learning models, while model in- terpretability was clearly superior for the BFE. The results highlight the technical performance of both the traditional and deep learning approaches and describe how they may translate to clinical utility. Furthermore, these algorithms can be applied to clinical datasets for novel prognostic and mechanistic biomarker discovery.

show abstract

Section: Glomerular Classificationmentioning

confidence: 99%

Computational Classification Methods in Renal Pathology

Basso

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The Segment Statistics module was used to compute tumor volume using a binary labelmap representation of the segment. To compute the shape, intensity, and texture features, 6 the radiomics extension (based on the PyRadiomics package in Python) was used. 7 Texture features quantify intratumor heterogeneity, which is an important factor that determines the prognosis.…”

Section: Feature Extractionmentioning

confidence: 99%

A Comparison of Machine Learning Models for Survival Prediction of Patients with Glioma Using Radiomic Features from MRI Scans

Manjunath

Saravanakumar

Kiran

et al. 2023

Indian J Radiol Imaging

View full text Add to dashboard Cite

Background Glioma is a primary, malignant, highly aggressive brain tumor, with patients having an average life expectancy of 14 to 16 months after diagnosis. Magnetic resonance imaging (MRI) scans of these patients can be used to extract and analyze quantifiable features with potential clinical significance. We hypothesize that there is a correlation between radiomic features extracted from MRI scans and survival. Along with clinical data, the radiomic features could be used in survival prediction of patients, providing beneficial information for clinicians to design personalized treatment plans. Methods In our study, we have utilized 3D Slicer for tumor segmentation and feature extraction and performed survival prediction of patients with glioma using four different machine learning models. Results and Conclusion Among the models compared, we have achieved a maximum prediction accuracy of 64.4% using the k-nearest neighbors model, which was trained and tested on a combination of clinical data and radiomic features extracted from MRI images provided in the BraTS 2020 dataset.

show abstract

“…Rathi et al [25] explore 60 features, including 22 shape, 5 intensity, and 33 texture features, for brain cancer detection. After selecting the best features using principal component analysis, their Support Vector Machine model achieves an accuracy of 0.98 in detecting malignant brain cancer in MRIs.…”

Section: Literature Surveymentioning

confidence: 99%

Explainable Convolutional Neural Networks for Brain Cancer Detection and Localisation

Mercaldo,

Brunese,

Martinelli

et al. 2023

Sensors

View full text Add to dashboard Cite

Brain cancer is widely recognised as one of the most aggressive types of tumors. In fact, approximately 70% of patients diagnosed with this malignant cancer do not survive. In this paper, we propose a method aimed to detect and localise brain cancer, starting from the analysis of magnetic resonance images. The proposed method exploits deep learning, in particular convolutional neural networks and class activation mapping, in order to provide explainability by highlighting the areas of the medical image related to brain cancer (from the model point of view). We evaluate the proposed method with 3000 magnetic resonances using a free available dataset. The results we obtained are encouraging. We reach an accuracy ranging from 97.83% to 99.67% in brain cancer detection by exploiting four different models: VGG16, ResNet50, Alex_Net, and MobileNet, thus showing the effectiveness of the proposed method.

show abstract

Brain tumor MRI image classification with feature selection and extraction using linear discriminant analysis

Cited by 7 publications

References 5 publications

Computational Classification Methods in Renal Pathology

Computational Classification Methods in Renal Pathology

A Comparison of Machine Learning Models for Survival Prediction of Patients with Glioma Using Radiomic Features from MRI Scans

Explainable Convolutional Neural Networks for Brain Cancer Detection and Localisation

Contact Info

Product

Resources

About